Humanitarian Intervention and Peacekeeping 8. The Ethics of Peacekeeping and Humanitarian Relief A Humanitarian Pragmatic Approach The Case of Zaire 1 Howard Adelman
I Preface Canada has taken a lead in assisting refugees 2, in peacekeeping, and in many other humanitarian endeavours in the international arena. This leadership has been particularly significant when Canadian initiatives are undertaken in response to complex emergencies. 3 All of these efforts are intended to help construct a better world. These initiatives stand in stark contrast to international initiatives presumably governed primarily by the self-interest of the intervening state and determined by realism instead of morality. 4 At the same time, these initiatives are intended to be practical and efficacious and are not simply the product of abstract norms or totally other-directed behaviour. This pragmatic idealism or humanitarian realism contrasts not only with realist-based behaviour and with strictly morally dictated behaviour, but also with ignorant behaviour in which neither norms nor rational calculation of interests are involved. Case studies of Canadian initiatives can be useful in understanding the ethical norms influencing such actions and if they were in conflict, how they were reconciled or mediated. We do not have to choose between an amoral realism and an abstract principled idealism. We really choose amongst competing norms in light of the circumstances as perceived and the anticipated consequences of each alternative. I imply which norms should or should not be operative in governing responses to international crises either politically or theoretically, and which norms are the relevant ones, only after I ask which norms were involved. What impact and role did they play? Then and only then can I ask what role should they have played in the complex emergency and the international response to it. Thus, the chapter will deal both with the connection between ethics and self-interest, including security concerns, as well as with the connection of these themes in the formation and critique of Canadian 2
foreign policy. Lessons learned, for Canada or more generally, can vary with the case used. For example, Raimo Väyrynen s introductory chapter, How Much Force in Humanitarian Intervention, takes Yugoslavia as a benchmark. Compared to Zaire, the Yugoslavian case is complicated by many more political, military, and economic considerations. It is a difficult case to serve as a benchmark to examine international ethics. I prefer the simpler one of Zaire 5, and not simply because Canada played such a prominent role in contrast to the Canadian participation in Yugoslavia. In Zaire, although other alternative options were available such as taking no action, or, at the other extreme, imposing a regime on the area through the use of overwhelming military force in fact, two distinct alternatives were considered. Either address the issue of the militants who controlled the camps or restrict activities to support for humanitarian operations. The latter might involve providing protection for the delivery and distribution of humanitarian relief supplies or also include protection for the refugees themselves in the camps or in flight. In former Yugoslavia, there were many more options while the capacity of the West to implement most of them was questionable. Yugoslavia was a minefield in modern ethical and political theory as well as on the ground. In Zaire, however difficult the choices, the options were clearer and there were fewer theoretical political issues at stake so that the problem of assessing the ethical issues is far less complicated. 6 II Background In addition to linking ethics with issues of self-interest in the context of the development of 3
Canadian foreign policy, I want to place the case within a temporal and spatial context. For crises, and the responses to them, are context dependent. What can and should be done in Asia or Europe is different than what can be done in Africa. Further, since the end of the Cold War, the context for dealing with issues of conflict in Africa has shifted dramatically. The response is no longer determined primarily as a by-product of East-West rivalry. A civil war had been fought in Rwanda from 1990 to 1994. The war ended with a victory of the rebels, but only after a genocide had taken place in which at least 500,000 Tutsi were killed. 7 The perpetrators were included among the fleeing masses. Hundreds of thousands of Hutu refugees fled to Zaire along with the military and militias who had been implicated in the genocide. The Hutu were not the intended victims of the genocide. Both the genuine Hutu refugees and the genocidaires feared reprisals from the Tutsi-led victorious rebel army in Rwanda, the Rwanda Patriotic Front (RPF). The refugee camps that were constructed were soon controlled by the militants who used them as bases to launch attacks back into Rwanda, to attack prisons to free those accused of genocide and to kill potential witnesses. When these strategies began to fail, the militants, concerned that they would be thrown out of Zaire as Mobutu was dying, tried to secure their place in Zaire by allying with local ethnic groups and attacking local Tutsi. Thus, the Rwanda civil war continued to be fought using refugee camps as bases. When local Tutsi became the targets, a new civil war was started in Zaire. Though 300,000 local Tutsi were killed or driven out in the Masisi region in North Kivu, when the genocidaires turned to attack the local Tutsi in South Kivu, the Banyamulenga were ready. They repulsed the attacks and, in turn, attacked the Hutu refugee camps, driving the refugees out. As the civil war developed and Kabila s Rwanda-backed forces eventually attacked the refugee camps from which 4
ethnic cleansing had been launched in the Masisi region, the camps were quickly evacuated by hundreds of thousands of fleeing refugees. Over 640,000 Hutu, freed from the intimidating presence of the genocidal militants, crossed the border back into Rwanda. The civil war moved west away from the Rwandan border. Previous to the defeat of the militants and the release of the refugee camps from their control, there had been many calls for the use of coercive force under UN auspices to disarm the ex-far (Forces Armées de Rwanda) and the interahamwe (Rwandese militia under the Habyarimana regime). The requests for intervention by UNHCR to disarm the militants started in 1994 not long after the camps were set up. 8 The debate resumed in March of 1996 when the refugee camps were used to attack the local Tutsi population. The discussion of options escalated considerably in the fall when the refugees began to flee the camps as the ADFL began to score victories. The concern was then not focused on removing the militants from control but on the use of military force to protect humanitarian workers and to ensure that water, food, shelter, and health services reached the refugees. The rebellion in Zaire, set off by the militants in the camps attacking the local Banyamulenge population, had cut the refugees off from the humanitarian agencies servicing the camps. The immediate crisis seemed to be their humanitarian plight. Many Africans viewed the proposed 1996 intervention that focused only on humanitarian aid as interference in a civil war that would drastically affect the rebels ability to prosecute the war. On the other hand, humanitarian intervention was defended as a necessity to fulfil the first obligation of agencies to provide aid to the refugees. Opponents of the plan argued that countries had not been interested in protecting the refugees with military force when those same refugees were being intimidated by the 5
extremist Hutu. Defenders of the plan countered that the lives of the refugees were not threatened in the camps in the same way as they seemed to be when the civil war broke out in Zaire in 1996. This position subsequently was viewed as fully justified given the reports of alleged atrocities against the refugees after all the camps had been evacuated. In November 1996, Canada launched an aborted peacekeeping mission with the intention of creating a militarily protected corridor to ensure that humanitarian relief (food, water, health services, etc.) could be delivered to the Hutu refugees in Zaire. 9 The primacy of humanitarian approaches over political ones seems to be a common characteristic of most of the attempts to deal with intrastate conflicts in the nineties. This was a product of a combination of elements prevalent after the end of the Cold War. When the communist threat had disappeared as either an inspiring idea or a real military threat on the ground, the significant force of humanitarian principles appeared from under the shadow of the realist paradigm that seemed to dominate in the Cold Ware era. Further, the predominance of the CNN effect had emerged. That is, the portrayal of humanitarian disasters on television stimulated in the public a demand that the government act to do something, even if the issue was not one that affected the self-interest of the state. Further, the legacy of the Cold War had left a left-leaning public oriented to believing that coercive military action risked initiating a nuclear war in which the whole world would be destroyed. They were wary of military adventures that could risk a nuclear war. The right-leaning public and the military were convinced that unless a war could be won quickly with massive use of force, public support of the military would vanish. The politicians would then abandon the effort to pursue and win the war leaving the army to suffer a humiliating defeat. As a result of the combination of both attitudes, the residue in the nineties was a low or no risk attitude to military intervention. There was a 6
demand for military humanitarianism but at no military risk. Even when self interest was at stake and intervention could be defended on humanitarian and legal grounds, as with the conquest of Kuwait by Iraq just when the Cold War ended, then massive force could be employed but still under the principle of no or little risk. At the same time, the Gulf War misled thinkers in understanding that a very different, though complementary, set of principles had emerged to influence the uses of military intervention. So the realist paradigm was seen as one at odds with rather than complementary to these principles. Further, the principles were caricatured as unrealistic bleeding heart moralism, sometimes with justification. III The Principles of Intervention Thus, though the Zaire case is relatively more straightforward than the crisis in former Yugoslavia, a number of principles common to both cases seemed to emerge. This list of principles does not purport to be exhaustive. They refer to the treatment of the victims of conflict, the relations with local states and prime actors in the conflict as well as the conduct of interveners from the international community. 1. Repatriation of Refugees: In contemporary doctrine, repatriation to the home country is viewed as the best solution for refugees. No consideration seemed to have been given to the possibility of resettlement abroad or the permanent settlement of the refugees in Zaire. 2. Voluntary Repatriation: 7
Voluntary repatriation for refugees was upheld as a dominant principle with respect to refugee return even though there was overwhelming evidence that the refugees were not free to choose to return even if they wanted to go home because of intimidation by the extremists. 10 3. Physical Protection for Refugees and the Internally Displaced: The international community is legally and morally obligated to provide physical and legal protection for refugees. There was some success in providing physical protection but little headway was made in protecting the rights of the refugees who were subject to the control and manipulation by the Hutu extremists. 4. Humanitarian Aid and Assistance: Providing humanitarian assistance to the refugees was not only a moral imperative, but this humanitarian imperative became the dominant governing principle for most NGOs and international agencies, and for many it was so predominant that it eliminated the consideration of other ethical imperatives. 5. Refugees and Refugee Warriors: International Refugee Law as well as the OAU Convention 11 require that refugee camps not be used to launch attacks against the countries from which the refugees fled and certainly not against local civilian populations. This principle was recognised but not enforced by the international community. 6. Respect for Sovereignty Consent of the parties, characteristic of classical peacekeeping, was a governing principle of any peace operation. However, the consent of the Zairian government was only nominal for the overall operation. The Zairian government was ignored in terms of obtaining permission for the entry point for 8
the peacekeepers or the advance military mission. That military mission communicated directly with the rebels without the permission of the Zairian government. 7. Political Impartiality The intervention was intended to be neutral. Part of the reason for the choice of Canada to lead the mission was to provide that image of impartiality as well as to paper over differences and leanings of major powers, more specifically, France and the United States. Further, the type of intervention and its purpose was regarded by a number of African states as itself partial. The Rwandans viewed the intervention as favouring the other side. 12 An intervention that did not confront the refugee warriors was unacceptable to the victorious rebels. The interposition of a foreign military army was seen as a way of preserving the status quo for the refugee warriors and inhibiting the prosecution of the civil war. 8. Financing Humanitarian Aid versus Humanitarian Intervention utilizing Peacekeepers: Major powers, particularly the US, are wary of even the financial burden of peacekeeping. This is especially true if there is a prospect that the peacekeepers could become embroiled in a civil war. Yet the failure to deal with the underlying political issues may be more costly in the long run, especially when a new round of fighting is merely postponed. But there seemed to be greater readiness to absorb the high cost of humanitarian aid and reconstruction after the conflict than the costs of military intervention to prevent the conflict. 9. Low Risk Military Humanitarian Intervention: International intervention is morally obligated, at the very least, in cases of genocide 13 : a) When the refugee men in one UNHCR (United Nations High Commission for Refugees) camp in the 9
former Yugoslavia were slaughtered when the camp was overrun by the enemy, the UN was unwilling to take effective action. b) When the militants in the Zairian camps attacked the local Banyamulenge population in Masisi in the spring of 1996 as a continuation of the genocide, no humanitarian intervention was even contemplated. This was the case even though the effective military capacity on the ground was very limited, certainly in comparison to Yugoslavia. A low or even NO risk approach to international peacekeeping seemed to be the governing norm. Thus, the feature characteristic of the Zaire case (and shared by the Yugoslav case) was the governing principle adopted: delivery of humanitarian aid was more important than a political or military solution. Further, military forces were required to ensure the delivery of that aid, as much as possible, indifferent to the merits or demerits of either of the contending sides. The international community maintained the right to the limited use of force for self-defence and humanitarian purposes. However, the potential of even this limited degree of military involvement served as a deterrent for military intervention at the same time as pressure for intervention increased to ensure the continued provision of humanitarian aid to the refugees. Ironically, to add weight to ensure both that the military would not be used and, if it was, there would be a limited number of casualties, the intervention when it was sanctioned, was authorized under a UN Chapter VII provision. This allowed a very activist militant approach to those undermining the humanitarian effort at the same time as it deterred any such effort. 10. Coherence amongst the Interveners. The real power brokers in the equation were the various countries involved in the issue. It has become a commonplace to assert that intervention should not take place unless the intervening parties 10
are governed by agreed upon goals and governing norms. But the decision to intervene was also a byproduct of the conflict among the major powers, specifically between the United States and France. France was really backing the Mobutu regime, and, therefore, in effect, backing the extremists in the camps, at least indirectly. The United States was giving support to Rwanda, and, through Rwanda effectively to the rebels, again, at least indirectly. The Americans, following PDD 25 14, were reluctant to become involved 15 and tended to want to leave the problem to local states, in particular, Rwanda and Uganda, to solve. Though for a short period in October 1996 the Americans deviated somewhat from this pattern, they quickly returned to their post-somalian norm of strictly limiting American involvement in humanitarian interventions. France, at the other extreme, was willing, if not eager, to land troops in Zaire to protect the refugees, and, hence, inhibit the advance of the rebels, thereby protecting Mobutu. Though approaching the problem from opposite standpoints, and though each country had a unique logistic capacity to airlift troops and equipment, both were opposed to becoming involved in disarming the extremists in the camps, and the US was a reluctant supporter of any humanitarian intervention. In that stance, they stood in total opposition to the regional African states who had volunteered to send troops to separate the militants from the rest of the refugees, but wanted an international sanction as well as financial and logistic support for such an initiative. IV Counting, Classifying and Accounting One does not normally think of numbers as involving ethics. But the number of refugees became a central ethical and political issue in Zaire. There were three ethical issues related to the number of 11
refugees accuracy, association and accountability. The basic ethical issue was straightforward. How many refugees were there? What seems like the most basic of factual issues becomes an ethical issue when we attach the norm requiring accuracy. The issue of association becomes clearer if we ask, How many genuine refugees were there? More specifically, how many genuine Rwandan Hutu refugees were there in the camps as distinct from Burundian refugees or militants? The issue of accountability focuses on the responsibility of agencies and states when they disseminate inaccurate and misleading figures. In other words, who was to blame for the confusion about the numbers? The variations in numbers had significant political effects. If there were 1,200,000 refugees and only 640,000 returned, then there were 560,000 refugees who fled west. 16 Since most did not reappear, they were allegedly slaughtered. This made the treatment of these refugees almost equivalent to the genocide in Rwanda. If the genocide in Rwanda obligated international intervention, so did the disaster in Zaire. In light of the failures in Rwanda, the plight of the refugees combined with a concern with the prevention and mitigation of another alleged genocide, the international community was obligated to intervene in Zaire to save the lives of the refugees. 17 Further, the American military was accused of trying to air brush the refugees out of history. 18 These are the central issues. They were rooted in what became known as the bataille des chiffres. What are the facts? How many refugees were there in Zaire in the first place? Purportedly, 1,200,000 19 refugees had fled to Zaire. A joint mission (UNHCR, WFP - World Food Program, USAID - United States Assistance for International Development, and Echo - the European Community Humanitarian Organization) estimated that there were 1,106,000 people. Note that the figure refers to people, not refugees. 20 These numbers in the camps in Zaire were before the upheavals 12
in the Fall of 1996 and the large return of refugees to Rwanda. 21 Of these refugees in Zaire, 140,000 were Burundians. 22 This meant that a total of 966,000 Rwandese refugees and refugee warriors were in Zaire, assuming the joint mission figures were correct. 23 But, as stated above, this figure, assuming that it was correct, included refugee warriors. Of the over one and one-half million Hutu Rwandese who fled Rwanda between April and July of 1994, 10 to 15 percent were alleged to have participated directly in (the) mass killing in Rwanda. The extremist militant group included the hard-line political leadership at all levels. Almost all of them fled to Zaire. Thus, of 966,000 Hutu Rwandans, 140,000 to 210,000 were genocidal killers or their families. 24 The estimate of the number of militants or refugee warriors can be approached another way. A report to the UN Secretary-General in 1994 (Degni-Ségui 1994, 16) divided the non-refugee Rwandan population in eastern Zaire into three groups: 1. Former leaders, principally consisting of 50 families lodged in Villas at Bukavu; 2. An estimated 16,000 military personnel of the ex-far (with families, the population of this group numbered 80,00); 3. The militants in the militia, possibly 50,000, but probably more like 35,000, and, in any case, difficult to enumerate since they lived amongst the refugees; including family members, since far fewer of them were accompanied by families compared to the ex-far, their numbers perhaps totaled around 100,000. This meant that approximately 180,000 refugee warriors and their family members were in the camps. Deducting the militants and their families from the realistic figure of 966,000 Hutu Rwandan refugees in the camps leaves approximately 786,000 genuine refugees in the camps - based on the 13
assumption that the original UNHCR figures were accurate. It is generally the case that camp populations are normally exaggerated by an average of 10%. and much more when controlled by militants. If so, then even these figures are inflated by at least 80,000 and possibly as much as 200,000. Assuming the lower figure, this would make the genuine Rwandan Hutu refugee population in Zaire just over 700,000. Of these, 15,000 were forcefully repatriated by the Zairian army in August of 1996. An additional 646,000 repatriated spontaneously in November. Assuming that only genuine refugees repatriated 670,000 of them this left 30,000 missing refugees among the approximately 180,000 militants and their families who fled west. If the Canadian 25 and American military calculated estimates of the camp population of only 900,000 (not 966,000) are accurate, then very few genuine refugees could have been killed. 26 The political issues were directly related to the above ethical issues. From the very beginning, the major issue was how to separate the criminals from innocent refugees. After the camps were attacked by the rebels and emptied, the mission to separate genuine refugee from the militants had already been accomplished by the rebel forces. Further, the case of the missing refugees seemed to be a phantom issue. So there was neither a protection nor a humanitarian issue. By then doubt about the mission was already widespread in the media Vancouver Sun front page 19 November, Globe and Mail front page 18 November, Toronto Star front page 18 November. The very next day the same papers headlined on the same front page that the mission was going forward. By 21 November, the Winnipeg Free Press (B1) and the Montreal Gazette (B1) were pronouncing the Canadian plan in disarray, stimulated, no doubt, by Chrétien s statement the day before that Canada might give aid rather than troops. 27 14
This issue of simple fact numbers became a critical factor in the debate over continuing the military mission after the 646,000 refugees had returned home in November 1996. Some UN and NGO spokespersons insisted on continuing the mission because 400,000 refugees were still missing and dying in the jungles of Zaire. For example, the UN Secretary-General himself said that, as of 18 November 1996, approximately 600,000 and close to 150,000 Burundian refugees remain in Zaire. 28 The news reports had already reported that 500,000 refugees had been repatriated. 29 UNHCR spokesperson, Melita Sunjik, issued a statement on 22 November 1996 that 700,000 refugees remained in eastern Zaire who had been located on satellite photos. 30 The January UN Secretary- General s Report tried to introduce some correction. It stated, on the one hand, that the number of returnees was underestimated by referring to 646,000 refugees as several hundred thousand. At the same time, the missing were reduced from hundreds of thousands to tens of thousands in a diplomatic balancing act that insisted that this does not mean that the refugee problem has been solved. At a minimum, several tens of thousands remain unaccounted for in Zaire, their whereabouts undiscovered and their living conditions unknown. It is clear, however, that every effort should continue to be made to locate them and provide them with food, shelter, and medicines to meet their humanitarian needs. 31 Hundreds of thousands had become tens of thousands, consistent with the belief of many that the only people really left in Zaire from Rwanda were the genocidaires, their families, and their captive Hutu carriers. The rest were phantom refugees. The military eventually told a different story. According to Canadian reports, the total number of refugees moving away from the border with Rwanda remained at about 200,000 when 640,000 refugees had already returned. The Canadian report further concluded that humanitarian access to those 15
refugees was available, and that the mission for which they were deployed had been accomplished, especially given the erosion in support in the region for the deployment of the Multinational Neutral Force (MNF). This confirmed MNF commander General Maurice Baril s assessment on 3 December that the MNF mission has largely been accomplished and the mandate should come to an end. More significantly, Baril noted that, some HRAs persist in their attempt to paint a continued humanitarian crisis in the making and deny the factual information made available to them at the local level. 32 The debate over such a basic fact as numbers had an important implication on what was to be done following the spontaneous return, a debate that should and could have been resolved but, instead, led to a great deal of animosity between peacekeepers and Non Governmental Organizations. 33 Who was to blame? The UNHCR had originally used inflated figures. But these were later corrected. Most NGOs, on the other hand, fixated on the highest possible figures and fed the media that disseminated these hysterical counts. Further, the NGOs went further and accused the American military of a coverup and the Canadian forces for being lap dogs of the Americans. But all NGOs cannot be tarred with the same brush. An MNF assessment report of 10 December 1996 stated, If one begins with the figure of 1.1 million and takes account of possible overestimates of up to 20% based on generous food distribution and over-registration, it is possible that the real number of refugees in Eastern Zaire at the beginning of Nov 96 was closer to 900,000. Some 640,000 are known to have returned to Rwanda in the past few weeks. If the figure of 200,000 is accepted as being a reasonable estimate of those who may not want to return then we would be able to account for 90% of the original estimate of the total refugee population. 34 The failure to resolve this most basic of debates not only made it much more difficult to arrive at 16
a policy for the missing refugees, but directly affected the credibility of the United Nations and the UNHCR. It also affected the image of who was to be held responsible for moral crimes against humanity. There was a second related problem one of classification that directly affected the issue of facts. Refugee warriors, 35 as distinct from genuine refugees, could have numbered about 180,000 if families of the militants are included. The camps had been controlled by ex-far (Forces Armées de Rwanda, the Rwandese army under Habyarimana) and interahamwe militia which largely had been responsible for the genocide. Thus, armed warriors, who are not by international law genuine refugees, and many of whom were likely criminals guilty of genocide who had launched military excursions into Rwanda and killed civilians, and had instigated conflict between local Hutus and the Bamyamulenge, were included under the designation refugee needing humanitarian assistance. The international community, which provided the aid to the refugee camps, had been impotent in separating the innocent refugees from their militant controllers to facilitate repatriation. They even failed to ensure that excess humanitarian aid was not purloined by the ex-far (Forces Armées de Rwanda) and sold on the black market, now compounded the problem by conveying an image of helpless deprived refugees fleeing westward when many were genocidal killers and militants. But this was not the view conveyed by many if not most of the highly respected international human rights bodies. The Commission estimates that based on various reports and testimonies of allegations approximately 200,000 refugees on Zairean soil, the majority of whom are ethnic Hutus, have lost their lives or disappeared in an arbitrary manner, 36 as a result of a deliberate strategy of gradual extermination of a portion of the Rwandan population. To this end, procedures were adopted, in a premeditated, constant, and persistent manner that strongly ressemble (sic!) acts of genocide. 37 As 17
Amnesty International had concluded earlier in the opening page of its report of 3 December 1997 called Democratic Republic of Congo: Deadly alliances in Congo forests, many of the more than 1,000,000 refugees from Rwanda and several hundred thousand from Burundi were being deliberately and arbitrarily killed in large numbers by forces of the main armed opposition group, the Alliance des forces démocratiques pour la libération du Congo (AFDL), Alliance of the Democratic Forces for the Liberation of Congo. 38 The confusion over basic facts such as numbers, and the categories for communicating those facts, had somehow turned the tables so that the killers suddenly became the victims of genocide, 39 an intentional strategy of the ex-far. V. Coherence Confusion over numbers and the categories under which those numbers should have been counted were not the only source of difference in determining what to do. Originally, two strategies for a peacekeeping force had been proposed. The first entailed directly addressing the issue of the militants who controlled the camps. There were two purposes for dealing with the issue of refugee warriors. They controlled and used genuine refugees for military, political and economic purposes in the pursuit of the conflict with the new government in Rwanda; this could have been stopped and the refugees could have been freed to make their own decisions. Secondly, the militants could have been prevented from attacking Tutsi (Banyamulenge) in Zaire, as they had in the Masisi district, and launching military excursions into Rwanda. In that way, not one but two wars could have been ended, and ended by dealing with the instigators of the war. There were other possible goals, such as bringing the genocidaires to justice, but this was not even considered. The second alternative entailed providing 18
protection for humanitarian relief supplies for the refugee camps and for the refugees who had fled the camps. Numerous parties had been urging the first policy, including the UNHCR and the head of USAID. 40 This was the recommendation of the summit of regional leaders held in Nairobi on 5 November 1966. However, the second option was adopted by the UN for the Canadian-led mission. Would an international intervention to take out the ex-far militants and interahamwe that controlled the camps have forestalled the attack by the rebels against the camps? If successful, would it have prevented the alleged disappearance of tens of thousands of refugees? Alternatively, was it more prudent to leave the freeing of the camps to the rebels, with the consequent return of the bulk of the refugees. This was clearly not the intent of those who advocated that a humanitarian intervention be restricted to the protection of humanitarian relief corridors to supply the refugees. The international community was in total disarray over what humanitarian use of coercive forces should involve. There were eight types of actors concerned with the refugees: a) refugee organizations claiming to represent the refugees in this case, such agencies were inseparable from the control of the ex-far and the interahamwe; b) the various ethnic groups involved the Congolese Tutsi or Banyumulenge, the Katangans, the Kasai, etc.; c ) various rebel groups in Uganda, Sudan, Angola, etc. who could forge alliances to obtain military training, arms, bases, and actions which could undermine the governments they were intent on overthrowing; d) humanitarian agencies delivering aid, food, and health care to the refugees; 19
e) international agencies with a prime concern with refugees (UNHCR) and the political situation (the UN itself and the Organization for African Unity); f) local states, particularly the states in which the refugees were located (Zaire) and the state from which the refugees fled (Rwanda and Burundi), but including Uganda, Tanzania, Kenya, Angola, and, given the regional implications of the crisis, Sudan, Ethiopia and Eritrea as well; g) overseas states with a concern for the refugees and/or the impending crisis in Zaire the USA, France, the EU, as well as countries such as Canada apparently more concerned with the humanitarian issues than the geo-political aspects of the crisis; h) military security services, mercenaries, local military and gendarmes, including forces of the Zairian army employed by the UNHCR for security purposes. These collective agents had different priorities, values, modes of operation and constituencies to which they are accountable. Refugee organizations represent the refugees but may not be accountable to them, particularly when militant political factions control the refugee camps. In the case of Zaire, the militants were both opposed to repatriation and certainly to any intervention in which they were the targets. Humanitarian agencies are concerned with the welfare rather than the political interests of the refugees, with their own organizations and the fundraising needed to help the refugees and sustain their organizations. On the issue of intervention, the NGOs were divided. A few supported the need to separate the militants from the genuine refugees. Others, while eventually conceding that military intervention was necessary or else the refugees might die, opposed the use of that military for anything but the security of humanitarian aid. International agencies, such as UNHCR, answer to their benefactors as well as their humanitarian mandates which restrict their activities to established modes 20
and objectives (such as voluntary return) and limit their abilities to provide security, while insisting that the prime functions of such an agency is protection for the refugees. In this case, UNHCR took a proactive and leading position in requesting security forces: a) to separate the militants from the rest of the refugees; and b) to protect the refugees and the aid workers. The intervention option was restricted to support for a purely humanitarian mission. In this context, when Canada offered its leadership and support, backing for the Canadian initiative only came from the great powers on condition that the mission had a restricted humanitarian mandate. This, of course, alienated the regional states in Africa who had not even been consulted on the decision. Though not predicted, it was predictable that these states, in particular Rwanda, would not co-operate with such an intervention. 41 What becomes clear is that our values are skewed. Volunteers go to these countries ostensibly to serve others. Peacekeepers go at some risk to themselves to mitigate conflicts in which they have little self-interest. But the actual priorities are protecting a state s reputation, protecting one s own soldiers, protecting international humanitarian workers, and, at the bottom, protecting the local population. The reality of this triage in security comes directly into conflict with the governing rhetoric. Further, there seems to be little recognition to the degree that this NO RISK self-interested policy is a product of the Cold War. Nor does there seem to be much understanding of how counter-productive it is. What is more, it allows policy makers to operate in a fantasy world that disregards local concerns, interests, and, especially goals. Without local co-operation, security services, even with a very restricted humanitarian mandate, were in no position to fulfil their mandate. Fortunately, the spontaneous return of over 600,000 refugees to Rwanda allowed the military mission to declare victory without 21
actually even being deployed on the ground. After all, the decision to deploy the troops was the catalyst that advanced the rebel timetable for attacking the camps. Nevertheless, innocent women and children were effectively abandoned, even if they were the wives and children of genocidaires and up to 30,000 genuine refugees forced to flee with them. We still do not know how many of them died and how they met their deaths, though there are too many substantiated rumours not to suspect that thousands were killed. In good part, the abandonment of these people must be attributed to the failure of the international community to adopt a coherent and effective policy for dealing with the refugee crisis. Within and among these groups there are many debates on how to co-operate and create coherent action, 42 but the inability to effect such coherence has had drastic effects on the security and welfare of the refugees, as in the Kibeho massacre. 43 This affects the determinations of how return is to be effected, the timing of any return, the modes and pace of return, the destinations for returnees, etc. In reading those debates one cannot help concluding that the stand an agency took on the policy debate, as well as the factual and categorical issues, depended as much if not more on the culture that agency represented than the objective data. The peacekeepers from leading states and many of the international agencies came down on one side of the debate while many NGOs and virtually all human rights organizations came down on the other side of the debate. 44 VI. Control, Repatriation and Peace Most refugee returns are spontaneous, 45 preceding or immediately following the signing of an 22
accord. 46 The main decision-makers in refugee repatriation are not always or even mainly the international brokers or the parties to the conflict. The refugees themselves are the key decision- makers provided they are able to make decisions free from coercive intimidation. Peace agreements rarely make provision for this fact, the calculus refugees make on relative risks, or for the suddenness of refugee movements. However, in some cases, refugees are forced to return to their home countries or other countries following a war 47 with no provision or international assistance. 48 In such cases, refugees are one result of the peace. At the end of 1996, the return of the Rwandese refugees had slowed to a trickle. The refugees claimed to be afraid of the treatment they would be accorded upon their return to Rwanda. For most humanitarian agencies, the rhetoric of fear, as well as the reluctance of the refugees to move, was in part attributable to the militants selection of the scouts who went back to Rwanda and then returned to the camps to report negatively on conditions at home. The reluctance to return was due, in part, to the intimidation of the refugees by the militants, but more likely the fear of return based on deeply embedded beliefs as well as actual incidents of returnees being targeted by militants in Rwanda. The initial repatriation of Tutsi refugees in 1994 followed this norm since the return was a product of the peace wrought, not through a peace agreement, but by the victory of the RPF (Rwanda Patriotic Front) in Rwanda. The flight of the Hutu into Zaire in 1994 was itself different from the 1959-64 original flight of the Tutsi in at least three respects. The massive exodus of the Hutu in 1994 followed the breach of a peace agreement, the Arusha Accords, by the extremist Hutus who were eventually defeated; flight followed that defeat. In 1959-64, flight also followed the defeat of the ruling Tutsis, but there was no peace accord that had been breached. Secondly, it is well known that if refugees are not 23
repatriated and satisfactorily integrated, they often metamorphose into refugee warriors, perpetuating a cycle of violence; this happened to the Tutsi refugees of 1959-64. 49 However, in Zaire from 1994 to 1996, the refugee warriors were warriors first and became refugees second when the defeated army and militias fled across the Rwandese border. Thirdly, in 1959-64, the international community ignored the Tutsi exodus. In 1994-96, the international community did not respond effectively to the genocide itself. But when the genocide was virtually over, and the genocidists, along with hundreds of other innocents, fled Rwanda, the humanitarian intervention of the international community was certainly swift and relatively effective. That humanitarian assistance backed by an international force, however, was not a sign of moral virtue, but one of failure. It signified opting to provide humanitarian assistance when expatriots from the donor countries were involved, but failure to intervene to protect hundreds of thousands of lives from genocidal massacres in Rwanda when western aid workers were not around. Nevertheless, the eventual repatriation of the Hutu from Zaire conformed to the norm it followed a military defeat, that of the militant extremist Hutu by the rebels in Zaire and was unanticipated, sudden, and spontaneous. Intervention in Rwanda appeared at first to involve some risk, though that risk turned out to be virtually nil. This low risk had been signalled by the non-resistance to the French Operation Turquoise, but perhaps only because the French never provided a significant threat to the genocidists and did not venture into the countryside to prevent the killings there. In fact, many interpreted the French intervention as a cover to allow the genocidists to escape. 50 In refugee camps, the soldiers provided the infrastructure support for relief and protection for the aid workers who were at very little risk. Thus, a humanitarian intervention restricted to a humanitarian mission, seemed without risk. Alternatively, any 24
action against the well-equipped and apparently well-trained ex-far and the interahamwe seemed to be very risky. On the one hand, this proved to be a gross overestimation given the lack of real resistance offered by the Hutu armed extremists to the rebels. On the other hand, there was a gross underestimation of the fire-power in the hands of the militants given the array of arms 51 and the number of soldiers 52 under the control of the militants. The most unique feature, however, of the Rwandese refugee plight in Zaire was that the attacks on the camps were the catalyst for the spontaneous return of over 600,000 refugees from Zaire to Rwanda, a spontaneous return that relieved the pressure for intervention. Nevertheless, there were still at least 150,000 to 200,000 extremists, their families and genuine refugees who fled westward, though, as I indicated earlier, some estimates went much higher. Whether there were more or less, whether or not there were phantom refugees as a product of erroneous and deliberately exaggerated counts in the first place, the fate of the remaining refugees remained a problem, but one that the dissolved effort in humanitarian intervention was unable to tackle. Further, this group that fled would pose a continuing security problem both for Zaire as well as for Rwanda. Whether dealing with security in the camps financed by the international community, repatriation as the primary goal of the international community, arms flows into the area, or the development of military armies among the refugees and in control of them, the international community ends up appearing as a paper tiger with very little control over the direction or pace of events. VII. Fundamental Ethical Clashes and Moral Consistency 25
Underpinning the institutional cultural clashes and incoherence, though not reducible to them, and overlaying the apparent impotence within the international community, are fundamental differences over values and their ranking. Should refugees have the right to move anywhere, or are they virtual prisoners within their welfare camps? This right to movement is also a right not to be moved, that is refouled or returned under pressure, though sometimes the refugees are induced, pressured or even forced to stay as refugees by coercive force. If not-so-gentle means of persuasion are adopted to effect a return cutting down food rations to induce movement, presumably in a context in which militants might have been preventing a free choice then the refugees have not really been free to stay or return. All this assumes that the state to which the refugees are destined to return is genuinely interested in taking them back. Thus, both the right to return and not to return clash with the need to find a permanent solution for the refugees. 53 This clash between rights and needs overlaps with a conflict between refugee rights and the principle of a state authority as the primary responsible agent for protecting its citizens or strangers on the territory of that state. In the interest of accelerating refugee return and preventing camps from breeding a new wave of refugee warriors, a more proactive policy on refugee return and more robust actions may be adopted, as was the case by the Tanzanian authorities following the massive return of the Hutu refugees from Zaire to Rwanda. An inverse situation is prevalent in Israel 54 and Republika Srbska; state sovereignty and a commitment to the primary national group may stand in the way of both return and an acknowledgement of refugee rights of return. In Macedonia, the right of a small country to protect itself from an overwhelming influx of refugees who could claim a right of asylum resulted in 26
another dimension of the conflict. Fortunately, this one was overcome when a third principle, that of burden sharing was brought into play and many of the refugees were relocated from Macedonia. How do you rank the state legitimate interest in security against international laws and conventions that are based on universal moral benchmarks like human rights and the rights of individuals to live in a state that provides protection? Physical threats to the lives of the refugees and even humanitarian workers has led some humanitarian agencies to push physical security higher up on their list of priorities and led them to adopt more realist oriented policies, including the use of peacekeepers to guard their own organizations and the refugees. Some of these organizations have evenly actively advocated humanitarian intervention. Protection is, thus, the primary issue - protection from whom, by whom, against what threat and to what end. 55 The issue of protection extends to times of flight, within camps and on return. Protection extends to local inhabitants as well as refugees. 56 The rights of refugees are not the only problem. The duty to disarm refugee warriors is a very different obligation. Further, the problem of human rights protection is greatly complicated by refugee flows. 57 Kumar argued that a human rights field operation (HRFOR - Human Rights Field Operation in Rwanda) should be incorporated into a regional approach that would focus on: (a) disarming refugee warriors; (b) separating those who are suspected of having committed violations of international humanitarian law from those who are not suspected; (c) providing an environment conducive to repatriation for refugees who want to return; and (d) policing borders to deter violent incursions. (Kumar, Krishna, ed. (1997), 77-78) UNHCR adds to that list a formal invitation to repatriate by the new government, establishing international tribunals to end a culture of impunity for those who committed crimes against humanity, and proper management of the repatriation 27
process by means of staging areas, transit camps, and relief centres. 58 The above suggests that the more basic problem is not the competing interest of states, or the conflict between state interests and individual human or refugee rights, or the failure of existing international regimes to be effective, or even the failure to observe international laws let alone very lofty international norms. These competing ethical grounds are seen to be at the root of the problem, but are not. They appear to be the problem only of ethics is presumed to be the implementation of abstract principles. For such a conception demands that we have a moral framework that can overarch the differences between different ethical theories. But there exists not overarching principle that can overcomes the differences between those who espouse different fundamental premises for determining international policy. Classical realists believe that the prime determinants of international affairs are self interest and power. Liberal international regime realists agree, but believe that international agreements, treaties and institutions can mitigate these interest and power conflicts. Grotians hold international lawto be the final arbiter in international affairs. While moralists desire that international action be rooted primarily in a concern for human rights. If the demand for coherence can only be established by constructing an overarching ethical theory that can resolve these differences, then we lack the necessary foundation for constructing effective actions. Proposals for humanitarian intervention are simply products and compromises for these competing perspectives. The result is that efforts in developing a consistent ethical basis for dealing with international humanitarian situations, and, in particular, in providing the normative grounds for humanitarian intervention which can be effective in mitigating disasters, flounder on the absence of a solid ethical base. We have division rather than a consistent foundation for behaviour. 28